BARRETT’S ESOPHAGUS 1052–3359/02 $15.00
.

00

ROLE OF ACID AND BILE

IN THE GENESIS OF
BARRETT’S ESOPHAGUS
Werner K.H. Kauer, MD, and Hubert J. Stein, MD

Gastroesophageal reflux disease is a common foregut disorder. It is

complicated by specialized columnar cell metaplasia in the distal portion
of the esophagus, the so-called Barrett’s esophagus, in approximately
10% to 15% of patients in whom increased esophageal exposure to
gastric or duodenal juice is documented.21, 34 The factors predisposing to
the development of Barrett’s esophagus with its known premalignant
potential still are debated. Loss of lower esophageal sphincter resistance
and compromised esophageal clearance set the stage for a prolonged
exposure to the refluxate of gastric and duodenal juice.27 The refluxed
juice, that is, acid and duodenal components, individually or in combina-
tion may contribute to columnar metaplasia in patients who have gastro-
esophageal reflux disease.32 Numerous studies have been conducted to
investigate the role of acid and bile in the genesis of Barrett’s esophagus.

ANIMAL STUDIES

Since Norman Barrett originally described the columnar epithelial

lining of the lower esophagus in 1950,2 the histogenesis of this condition
has been disputed. After several clinical studies that suggested that
excessive gastroesophageal reflux disease is responsible for the develop-
ment of Barrett’s esophagus, investigators tried to support this by ani-
mal studies.

From the Department of Surgery, Klinikum rechts der Isar der TU München, München,
Germany

CHEST SURGERY CLINICS OF NORTH AMERICA

VOLUME 12 • NUMBER 1 • FEBRUARY 2002 39

40 KAUER & STEIN

Using a dog model, Bremner et al5 were the first to demonstrate that
a columnar epithelial metaplasia in the distal esophagus could result
from prolonged reflux of acid. This finding was confirmed by Gillen et
al,9 who studied canine esophageal mucosa under basal conditions and
in the presence of gastroesophageal reflux.9 Under normal conditions
mucosal defects in the esophagus are regenerated by squamous epithe-
lium. In the presence of gastroesophageal reflux of acid or a combination
of acid and bile, regeneration was frequently by columnar epithelium,
but this was not seen with bile reflux alone.
Most of the current knowledge regarding the noxious component in
the refluxed juice is based on the studies of Harmon and Johnson.11, 12
Using an in vivo rabbit model, they demonstrated that hydrogen-ion
injury to the mucosal barrier occurs mainly at a pH of less than 2. It
results in injury to the mucosal barrier but rarely produces mucosal
lesions or inflammatory changes. In an acid environment, the enzyme
pepsin seems the major injurious agent.
Lillimoe et al19 have shown that the reflux of bile and pancreatic
enzymes into the stomach can protect or augment esophageal mucosal
injury. In a patient whose gastric acid secretion maintained an acid
environment, the presence of bile salts would attenuate the injurious
effect of pepsin, and the acid gastric environment would inactivate
trypsin. Such a patient would have bile-containing acid gastric juice that,
when refluxed into the esophagus, would injure the mucosal barrier and
the epithelium but would be less caustic than the reflux of acid gastric
juice alone. In contrast, in a patient who has significant duodenogastric
reflux, a more alkaline intragastric pH environment may be present and
encourage the solubility of bile salts with a high pKa.
For bile acids to injure mucosal cells, they must be soluble and un-
ionized, so that the un-ionized nonpolar form may enter mucosal cells.10, 22
Before secretion into bile, 98% of bile acids are conjugated with taurine
or glycine in a ratio of 3:1. Conjugation increases the solubility and
ionization of bile acids by lowering their pKa.13 At the normal duodenal
pH of approximately 7, more than 90% of bile salts are in solution and
are ionized completely. At pH ranges of 2 to 7, a mixture of the ionized
salt and the lipophilic, nonionized bile acids is present.25 Acidification
of bile to a pH of less than 2 results in an irreversible bile acid precipita-
tion.3 Consequently, under normal physiologic conditions, bile acids
precipitate and are of minimal effect when an acidic gastric environment
exists. Conversely, in a more alkaline gastric environment, such as after
gastrectomy or with medical acid suppression therapy, bile salts remain
in solution, are partially dissociated, and when refluxed into the esopha-
gus, can cause severe mucosal injury by crossing the cell membrane and
damaging the mitochondria.18
This hypothesis is supported by a study by Ireland et al,14 who
manipulated rats so that the esophagus was exposed to reflux of gastric
juice, duodenal juice, or a combination of gastric and duodenal juice.
In this rat model, the presence of gastric juice protected against the
development of esophageal adenocarcinoma. The absence of gastric juice
 ROLE OF ACID AND BILE IN THE GENESIS OF BARRETT’S ESOPHAGUS 41

resulted in a threefold increase in the prevalence of adenocarcinoma.14

The protective effect of the stomach seems related to the secretion of
acid because there was a progressive increase in the prevalence of
esophageal adenocarcinoma as the amount of gastric acid that was
permitted to reflux with duodenal juice into the esophagus was reduced.

HUMAN STUDIES

Employing Esophageal pH Monitoring

Ambulatory 24-hour esophageal pH monitoring has become the

gold standard in the diagnosis of gastroesophageal reflux disease. Sys-
tematic studies showed that compared with patients who have increased
esophageal acid exposure but no columnar metaplasia, the quality and
the quantity of the refluxed juice seem different in patients who have
Barrett’s esophagus. On pH monitoring, this is reflected in a significantly
increased frequency and duration of reflux episodes with a pH ⬍4
compared with patients with reflux who have no columnar metapla-
sia.27, 29
Increased gastric acid secretion may contribute to the development
of columnar metaplasia by exposing the esophageal mucosa to a higher
degree of acidity. In patients who have Barrett’s esophagus, ambulatory
24-hour esophageal pH monitoring shows an increased esophageal expo-
sure not only to pH ⬍4 but also to more caustic gastric acid, that is, pH
⬍3 and pH ⬍2.27, 29
In addition to a significantly increased acid exposure, patients who
have Barrett’s esophagus also have an increased esophageal exposure to
alkalinity, as indicated by the time pH ⬎7 on esophageal pH monitor-
ing.28 The alkaline component of the refluxed juice seems to result
from a contamination of the refluxed gastric contents with excessive
duodenogastric reflux.1 Measurement of esophageal exposure to duode-
nal contents with pH monitoring is, however, less dependable than the
measurement of esophageal acid exposure.30 The significance of an alka-
line pH on prolonged esophageal pH monitoring as an indicator for
duodenogastroesophageal reflux therefore has been debated.

Employing Esophageal Bilirubin Monitoring

For reliable detection of duodenal contents in refluxed gastric juice,

a fiber-optic system for circadian monitoring of duodenogastroesopha-
geal reflux was developed by Bechi et al.4 With the Bilitec system
(Synectics Medical, Stockholm, Sweden) and by use of bilirubin as a
marker for duodenal juice, it has been shown that patients who have
reflux of acid gastric juice alone have less severe esophageal mucosal
injury than patients who have reflux of gastric juice contaminated with
duodenal components.15, 31, 32 Further, duodenal juice reflux into the
esophagus is significantly more common in patients who have Barrett’s
42 KAUER & STEIN

esophagus compared with patients who have erosive esophagitis or

patients with reflux who have no mucosal injury. In addition, the mean
percentage time of esophageal exposure to duodenal juice is significantly
higher in patients who have Barrett’s esophagus.15, 31, 32
Simultaneous esophageal pH and bilirubin monitoring showed that
esophageal exposure to duodenal juice occurs at all pH values.15, 17 In
patients who have gastroesophageal reflux disease, duodenal content
was detected within the esophagus 15% of the time when the pH was
less than 4, 19% of the time when the pH was between 4 and 7, and 6%
of the time when the pH was more than 7. An analysis of the cumulative
period during which the esophagus was exposed to duodenal juice
showed that the pH of the esophagus was between 4 and 7 more than
87% of the time. This pH is considered normal for the esophagus,
and consequently such reflux goes undetected and unappreciated when
analyzed by traditional criteria.
The Bilitec technology, although not a quantitative measurement of
bile reflux,20 showed that 58% of the patients who have gastroesophageal
reflux disease had increased esophageal exposure to duodenal juice, that
this exposure occurs most commonly when the esophageal pH is be-
tween 4 and 7, and that it is associated with severe esophageal mucosal
injury.17

Using Esophageal Aspiration Techniques

Although numerous studies have suggested excessive reflux of duo-

denal contents into the esophagus in patients who have Barrett’s esopha-
gus, few have measured this directly. Using prolonged ambulatory aspi-
ration in the distal esophagus it could be shown that patients who have
gastroesophageal reflux disease and Barrett’s esophagus have greater
and more concentrated bile acid exposure to the esophageal mucosa
than do normal subjects. This increased exposure occurs most commonly
during the supine period while asleep and during the upright period
after meals.16, 28 Aspiration studies also delivered more details on the
noxious effects of specific bile salts and their form, that is, conjugated
or deconjugated. Under normal circumstances deconjugated bile acids
are rare in the upper gastrointestinal tract. This rarity is because the
bacteria normally responsible for deconjugation are not present in an
acid environment. The authors identified the glycine conjugates of cholic,
deoxycholic, and chenodeoxycholic acids as the predominant bile acids
aspirated from the esophagus of patients who have gastroesophageal
reflux disease. This predominance is, as would be expected, because
glycine conjugates are three times more prevalent than taurine conju-
gates in normal human bile.

IMMUNOHISTOCHEMICAL STUDIES

Cell differentiation as determined by villin expression and cell pro-

liferation as determined by thymidine incorporation were investigated
 ROLE OF ACID AND BILE IN THE GENESIS OF BARRETT’S ESOPHAGUS 43

by Fitzgerald et al.8 They demonstrated in an ex vivo experiment that

acid influences cell proliferation and differentiation in Barrett’s esopha-
gus and that such an effect depends on the pattern of acid exposure.
Continuous acid exposure resulted in increased villin expression and
reduced cell proliferation, suggesting a differentiated phenotype. In con-
trast, after a short acid-pulse a fairly undifferentiated phenotype was
found.
The same group studied cyclo-oxygenase (COX)-2, which has been
shown to be involved in chronic inflammation and epithelial cell
growth.6, 23 Recent evidence has implicated COX-2 in colorectal, gastric,
and esophageal carcinogenesis.7, 26, 33
The role of COX-2 in various stages of Barrett’s esophageal metapla-
sia and in response to pulses of acid and bile salts in an ex vivo organ
culture system was investigated.24 There was a progressive increase in
expression of COX-2 with disease progression from Barrett’s metaplasia
to dysplasia and adenocarcinoma. This increase indicates that COX-2
overexpression is an early event in the neoplastic transformation process
of Barrett’s columnar metaplasia. Even more interestingly, these studies
showed that bile and acid could induce COX-2 expression in ex vivo
human epithelial explants, because COX-2 induction was increased sig-
nificantly in the presence of acid and bile. The highest induction could
be found when the explants were exposed to a 1-hour pulse of bile
salts, which, in part, could be related to protein kinase C activation by
bile salts.8

SUMMARY

Clinical and experimental studies have shown that acid and bile
reflux are increased in patients who have Barrett’s esophagus. The com-
bination of both seems the key factor in the pathogenesis of Barrett’s
esophagus. This factor has been confirmed by immunohistochemical
studies that show that environmental factors, such as acid and bile, are
involved in the pathogenesis of Barrett’s esophagus.
There is a critical pH range between 3 and 6 in which bile acids
exist in their soluble, un-ionized form; can penetrate cell membranes;
and accumulate within mucosal cells. At a lower pH, bile acids are
precipitated, and at a higher pH, bile acids exist in their noninjurious
ionized form. Thus incomplete gastric acid suppression, as is the case
with most medical treatment regimens for gastroesophageal reflux, may
in fact predispose to the development of Barrett’s esophagus.32

References

1. Attwood S, DeMeester T, Bremner C, et al: Alkaline gastroesophageal reflux: Implica-

tions in the development of complications in Barrett’s columnar-lined lower esophagus.
Surgery 106:764–770, 1989
44 KAUER & STEIN

2. Barrett N: Chronic peptic ulcer of the esophagus and esophagitis. Br J Surg 38:175–
182, 1950
3. Barthlen W, Liebermann-Meffert D, Feussner H, et al: Influence of pH on bile acid
concentration in human, pigs and commercial bile: Implications for ‘‘alkaline’’ gastro-
esophageal reflux. Dis Esophagus 7:127–130, 1994
4. Bechi P, Falciai R, Baldini F, et al: A new fiberoptic sensor for ambulatory reflux
monitoring. Validation of a new fiberoptic technique. Dig Dis Sci 38:1297–1306, 1993
5. Bremner C, Lynch V, Ellis F: Barrett’s esophagus: Congenital or acquired? An experi-
mental study of esophageal mucosal regeneration in the dog. Surgery 68:209–216, 1970
6. DuBois R, Abramson S, Crofford L, et al: Cyclooxygenase in biology and disease.
FASEB J 12:1063–1073, 1998
7. Eberhart C, Coffey R, Radhika A, et al: Up-regulation of cyclooxygenase-2 gene
expression in human colorectal adenomas and adenocarcinomas. Gastroenterology
107:1183–1188, 1994
8. Fitzgerald R, Omary M, Triadafilopoulos G: Dynamic effects of acid on Barrett’s
esophagus. J Clin Invest 98:2120–2128, 1996
9. Gillen P, Keeling P, Byrne P, et al: Experimental columnar metaplasia in the canine
esophagus. Br J Surg 75:113–114, 1988
10. Harmon J, Bass B, Batzri S: Alkaline reflux gastritis: Mechanism of injury. Probl Gen
Surg 10:201–206, 1993
11. Harmon J, Doong T, Gadacz T: Bile acids are not equally damaging to the gastric
mucosa. Surgery 7:79–86, 1978
12. Harmon J, Johnson L, Maydonovitch C: Effects of acid and bile salts on the rabbit
esophageal mucosa. Dig Dis Sci 26:65–72, 1981
13. Heaton K: Bile salts. In Wright R, Alberti K, Karran S, et al (eds): Liver and Biliary
Disease. London, WB Saunders, 1979, pp 233–254
14. Ireland A, Peters J, Smyrk T, et al: Gastric juice protects against the development of
esophageal adenocarcinoma in the rat. Ann Surg 224:358–371, 1996
15. Kauer W, Burdiles P, Ireland A, et al: Does duodenal juice reflux into the esophagus
of patients with complicated GERD? Evaluation of a fiberoptic sensor for bilirubin.
Am J Surg 169:98–104, 1995
16. Kauer W, Peters J, DeMeester T, et al: Composition and concentration of bile acid
reflux into the esophagus of patients with gastroesophageal reflux disease. Surgery
122:874–881, 1997
17. Kauer W, Peters J, DeMeester T, et al: Mixed reflux of gastric and duodenal juice is
more harmful to the esophagus than gastric juice alone. Ann Surg 4:525–533, 1995
18. Krahenbuhl S, Talos C, Fischer S, et al: Toxicity of the bile acids in the electron
transport chain of isolated rat liver mitochondria. Hepatology 19:471–479, 1994
19. Lillimoe K, Johnson L, Harmon J: Alkaline esophagitis: A comparison of the ability of
components of gastroesophageal contents to injure the rabbit esophagus. Gastroenterol-
ogy 85:621–628, 1983
20. Ostrow J: Bilitec to quantitate duodenogastric reflux: Is it valid? Gastroenterology
108:1332–1334, 1995
21. Sarr M, Hamilton F, Marone G, et al: Barrett’s esophagus: Its prevalence and association
with symptoms of gastroesophageal reflux. Am J Surg 149:187–193, 1985
22. Schweitzer E, Bass B, Batzri S, et al: Bile acid accumulation by rabbit esophageal
mucosa. Dig Dis Sci 31:1105–1113, 1986
23. Seibert K, Zhang Y, Leahy K, et al: Pharmacological and biochemical demonstration of
the role of cyclooxygenase-2 in inflammation and pain. Proc Natl Acad Sci U S A
91:12013–12017, 1994
24. Shirvani V, Rodica O, Baljeet S, et al: Cyclooxygenase 2 expression in Barrett’s esopha-
gus and adenocarcinoma: Ex vivo induction by bile salts and acid exposure. Gastroen-
terology 118:487–496, 2000
25. Small D: The physical chemistry of cholanic acids. In Nair PP, Kritchecsky D (eds):
The Bile Acids: Chemistry, Physiology and Metabolism, vol 1. New York, Plenum
Press, 1971, pp 249–356
26. Soydan A, Gaffem J, Weech P, et al: Cytosolic phospholipase A2, cyclooxygenases and
arachidonate in human stomach tumors. Eur J Cancer 33:1508–1512, 1997
 ROLE OF ACID AND BILE IN THE GENESIS OF BARRETT’S ESOPHAGUS 45

27. Stein HJ, Barlow AP, DeMeester TR, et al: Complications of gastroesophageal reflux
disease: Role of the lower esophageal sphincter, esophageal acid/alkaline exposure,
and duodenogastric reflux. Ann Surg 216:35–43, 1992
28. Stein HJ, Feussner H, Kauer W, et al: ‘‘Alkaline’’ gastroesophageal reflux: Assessment
by ambulatory esophageal aspiration and pH monitoring. Am J Surg 167:163–168, 1994
29. Stein H, Hoeft S, DeMeester T: Functional foregut abnormalities in Barrett’s esophagus.
J Thorac Cardiovasc Surg 105:107–111, 1993
30. Stein HJ: Characterization of acid and alkaline reflux in patients with Barrett’s esopha-
gus. Dis Esophagus 10:22–23, 1997
31. Stein HJ, Kauer WKH, Feussner H, et al: Bile reflux in benign and malignant Barrett’s
esophagus. Effect of medical acid suppression and fundoplication. J Gastrointest Surg
2:333–341, 1998
32. Stein HJ, Kauer WKH, Feussner H, et al: Bile acids as components of the duodenogas-
tric refluxate: Detection, relationship to bilirubin, mechanism of injury and clinical
relevance. Hepatogastroenterology 46:66–67, 1999
33. Wilson K, Fu S, Ramajunam K, et al: Increased expression of inducible nitric oxide
synthase and cyclooxygenase-2 in Barrett’s esophagus and associated adenocarcino-
mas. Cancer Res 58:2929–2934, 1998
34. Winters C, Spurling T, Chobanian S, et al: Barrett’s esophagus: A prevalent, occult
complication of gastroesophageal reflux disease. Gastroenterology 92:118–124, 1987

Address reprint requests to

Hubert J. Stein
Chirurgische Klinik und Poliklinik
Klinikum rechts der Isar der TU München
Ismaningerstr. 22
J-81675 München
Germany

e-mail: stein@nt1.chir.med.tu-muenchen.de